McKinsey reports on artificial intelligence get quoted in boardrooms and news feeds. They shape how leaders think about jobs, skills, and automation. But few readers know where this research comes from or how to weigh it. This guide explains both.
The short version: the research arm is real, large, and well funded. It also has a track record worth studying. Knowing that history makes you a sharper reader of the next big AI headline. For related coverage, see The Arabic large language models: Falcon, Jais, ALLaM.
Where the Research Comes From
McKinsey & Company, founded in 1926, is a strategy and management consulting firm. According to its reference profile on Wikipedia, consultants there regularly publish books, research, and articles on business and management. The firm spends $50 to $100 million a year on research.
Two outlets matter most. The McKinsey Quarterly, a business magazine, has run since 1964. The McKinsey Global Institute, founded in 1990, studies global economic trends. The profile adds that these publications give the firm a quasi-academic image.
The reach of the research mirrors the reach of the firm. In 2018, the profile records, 800,000 candidates applied for 8,000 jobs. Many recruits hold advanced degrees in science, medicine, engineering, or law, not just business. That mix shapes the style of the reports: data heavy, framework driven, and written for executives.
What It Says About AI and Work
AI now shows up across the firm's published work. Its September 2024 energy report pointed to rising electricity demand driven by artificial intelligence, a sign of how central AI has become to long-range planning. Research like this treats AI less as a gadget and more as a force that reshapes tasks, energy use, and business models.
For readers, the useful lens is simple. Ask which parts of work the study covers, such as tasks, hiring, or energy demand. Then ask what data stands behind each claim. Big firms publish method notes for a reason.
A Track Record Worth Weighing
Consulting research has had famous misses. The profile records an 1980s episode in which AT&T cut investments in cell towers after a McKinsey prediction of only 900,000 cell phone subscribers by 2000. The real number was about 109 million. A book quoted in the profile called that forecast laughably off the mark.
The profile also notes criticism of past work. A 1984 analysis by BusinessWeek found that many companies praised in the book In Search of Excellence no longer met the criteria two years later. None of this makes AI research wrong. It just means every forecast deserves a second look.
Talent research offers another lesson. A 1997 article and a 2001 book on the war for talent urged firms to rank staff and promote stars. The profile notes that Enron followed many of those principles before its collapse. Confident titles can rest on shaky ground.
How to Read the Next Report
Start with the scope. Was the study global or regional? Which industries were surveyed? Next, look for the base year behind any number. Finally, separate what the data shows from what the authors project. Both matter, but they are not the same thing. For example, note whether a survey covers all employers or only large ones. Coverage gaps are where most misreadings start.
It also helps to compare findings across outlets. If a claim appears only in one consulting report, treat it as one view, not a settled fact.
Conclusion: Useful, But Read It Actively
McKinsey's research on AI and work is influential because it is broad, funded, and consistent. That influence is a reason to read it carefully, not a reason to skip it. Treat the reports as informed starting points. This connects to our earlier piece, How to Read AI News Without Getting Fooled.
The best readers hold two ideas at once. The research can be serious work and still miss the future. Forecasts about AI and jobs deserve exactly that kind of steady, skeptical attention.
